| { | |
| "best_metric": 0.9358813166618347, | |
| "best_model_checkpoint": "/media/mldrive/kcardenas/limb_classification_person_crop/beit-large-patch16-384/8_1e-5_5e-4_0.15/checkpoint-1926", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 2140, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11682242990654206, | |
| "grad_norm": 1232665.625, | |
| "learning_rate": 5.000000000000001e-07, | |
| "loss": 1.603, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.2336448598130841, | |
| "grad_norm": 1125859.125, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 1.5674, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.35046728971962615, | |
| "grad_norm": 1237422.625, | |
| "learning_rate": 1.5e-06, | |
| "loss": 1.5855, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.4672897196261682, | |
| "grad_norm": 1008492.875, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 1.5179, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5841121495327103, | |
| "grad_norm": 825557.9375, | |
| "learning_rate": 2.5e-06, | |
| "loss": 1.4596, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.7009345794392523, | |
| "grad_norm": 461811.875, | |
| "learning_rate": 3e-06, | |
| "loss": 1.4276, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.8177570093457944, | |
| "grad_norm": 785723.125, | |
| "learning_rate": 3.5e-06, | |
| "loss": 1.3977, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.9345794392523364, | |
| "grad_norm": 589829.1875, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.4012, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.2777777777777778, | |
| "eval_loss": 1.3970400094985962, | |
| "eval_runtime": 14.9894, | |
| "eval_samples_per_second": 80.457, | |
| "eval_steps_per_second": 2.535, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 1.0514018691588785, | |
| "grad_norm": 396606.21875, | |
| "learning_rate": 4.5e-06, | |
| "loss": 1.3881, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.1682242990654206, | |
| "grad_norm": 353690.125, | |
| "learning_rate": 5e-06, | |
| "loss": 1.3671, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.2850467289719627, | |
| "grad_norm": 419103.875, | |
| "learning_rate": 5.500000000000001e-06, | |
| "loss": 1.3582, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.4018691588785046, | |
| "grad_norm": 374088.375, | |
| "learning_rate": 6e-06, | |
| "loss": 1.3592, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.5186915887850467, | |
| "grad_norm": 600778.9375, | |
| "learning_rate": 6.5000000000000004e-06, | |
| "loss": 1.3563, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.6355140186915889, | |
| "grad_norm": 5453174.5, | |
| "learning_rate": 7e-06, | |
| "loss": 1.3424, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.7523364485981308, | |
| "grad_norm": 657756.375, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 1.355, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.8691588785046729, | |
| "grad_norm": 560229.8125, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.3133, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.985981308411215, | |
| "grad_norm": 596723.375, | |
| "learning_rate": 8.5e-06, | |
| "loss": 1.2956, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.4709784411276949, | |
| "eval_loss": 1.2437535524368286, | |
| "eval_runtime": 14.8418, | |
| "eval_samples_per_second": 81.257, | |
| "eval_steps_per_second": 2.56, | |
| "step": 428 | |
| }, | |
| { | |
| "epoch": 2.102803738317757, | |
| "grad_norm": 661517.125, | |
| "learning_rate": 9e-06, | |
| "loss": 1.278, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.2196261682242993, | |
| "grad_norm": 499854.875, | |
| "learning_rate": 9.5e-06, | |
| "loss": 1.2474, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 2.336448598130841, | |
| "grad_norm": 353308.375, | |
| "learning_rate": 1e-05, | |
| "loss": 1.2492, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.453271028037383, | |
| "grad_norm": 568008.8125, | |
| "learning_rate": 9.847560975609756e-06, | |
| "loss": 1.2333, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 2.5700934579439254, | |
| "grad_norm": 377403.0625, | |
| "learning_rate": 9.695121951219513e-06, | |
| "loss": 1.2086, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.6869158878504673, | |
| "grad_norm": 376984.78125, | |
| "learning_rate": 9.542682926829268e-06, | |
| "loss": 1.2084, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 2.803738317757009, | |
| "grad_norm": 424205.8125, | |
| "learning_rate": 9.390243902439025e-06, | |
| "loss": 1.2112, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.9205607476635516, | |
| "grad_norm": 518474.75, | |
| "learning_rate": 9.237804878048782e-06, | |
| "loss": 1.1902, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5895522388059702, | |
| "eval_loss": 1.0810370445251465, | |
| "eval_runtime": 15.4574, | |
| "eval_samples_per_second": 78.021, | |
| "eval_steps_per_second": 2.458, | |
| "step": 642 | |
| }, | |
| { | |
| "epoch": 3.0373831775700935, | |
| "grad_norm": 363418.0625, | |
| "learning_rate": 9.085365853658538e-06, | |
| "loss": 1.1903, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 3.1542056074766354, | |
| "grad_norm": 472692.21875, | |
| "learning_rate": 8.932926829268293e-06, | |
| "loss": 1.1369, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 3.2710280373831777, | |
| "grad_norm": 389273.5, | |
| "learning_rate": 8.78048780487805e-06, | |
| "loss": 1.1591, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.3878504672897196, | |
| "grad_norm": 536079.4375, | |
| "learning_rate": 8.628048780487805e-06, | |
| "loss": 1.1742, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 3.5046728971962615, | |
| "grad_norm": 369198.15625, | |
| "learning_rate": 8.475609756097562e-06, | |
| "loss": 1.157, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 3.621495327102804, | |
| "grad_norm": 459633.09375, | |
| "learning_rate": 8.323170731707317e-06, | |
| "loss": 1.1494, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 3.7383177570093458, | |
| "grad_norm": 457643.6875, | |
| "learning_rate": 8.170731707317073e-06, | |
| "loss": 1.1559, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.8551401869158877, | |
| "grad_norm": 459124.59375, | |
| "learning_rate": 8.01829268292683e-06, | |
| "loss": 1.1376, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 3.97196261682243, | |
| "grad_norm": 416160.96875, | |
| "learning_rate": 7.865853658536587e-06, | |
| "loss": 1.1316, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6144278606965174, | |
| "eval_loss": 1.085984230041504, | |
| "eval_runtime": 15.1709, | |
| "eval_samples_per_second": 79.494, | |
| "eval_steps_per_second": 2.505, | |
| "step": 856 | |
| }, | |
| { | |
| "epoch": 4.088785046728972, | |
| "grad_norm": 415968.6875, | |
| "learning_rate": 7.713414634146342e-06, | |
| "loss": 1.1317, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 4.205607476635514, | |
| "grad_norm": 417537.21875, | |
| "learning_rate": 7.560975609756098e-06, | |
| "loss": 1.1403, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 4.322429906542056, | |
| "grad_norm": 560744.375, | |
| "learning_rate": 7.408536585365854e-06, | |
| "loss": 1.1016, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 4.4392523364485985, | |
| "grad_norm": 597802.5, | |
| "learning_rate": 7.25609756097561e-06, | |
| "loss": 1.0959, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 4.55607476635514, | |
| "grad_norm": 442672.09375, | |
| "learning_rate": 7.1036585365853665e-06, | |
| "loss": 1.103, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 4.672897196261682, | |
| "grad_norm": 427007.0625, | |
| "learning_rate": 6.951219512195122e-06, | |
| "loss": 1.092, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.789719626168225, | |
| "grad_norm": 545753.3125, | |
| "learning_rate": 6.798780487804879e-06, | |
| "loss": 1.0949, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 4.906542056074766, | |
| "grad_norm": 332691.28125, | |
| "learning_rate": 6.646341463414635e-06, | |
| "loss": 1.1133, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6475953565505804, | |
| "eval_loss": 1.0152363777160645, | |
| "eval_runtime": 15.3737, | |
| "eval_samples_per_second": 78.446, | |
| "eval_steps_per_second": 2.472, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 5.0233644859813085, | |
| "grad_norm": 608418.625, | |
| "learning_rate": 6.493902439024391e-06, | |
| "loss": 1.1168, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 5.140186915887851, | |
| "grad_norm": 500188.65625, | |
| "learning_rate": 6.341463414634147e-06, | |
| "loss": 1.0857, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 5.257009345794392, | |
| "grad_norm": 472268.6875, | |
| "learning_rate": 6.189024390243903e-06, | |
| "loss": 1.0784, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 5.373831775700935, | |
| "grad_norm": 512482.09375, | |
| "learning_rate": 6.0365853658536585e-06, | |
| "loss": 1.0912, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 5.490654205607477, | |
| "grad_norm": 553851.8125, | |
| "learning_rate": 5.884146341463415e-06, | |
| "loss": 1.0615, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 5.607476635514018, | |
| "grad_norm": 362454.03125, | |
| "learning_rate": 5.731707317073171e-06, | |
| "loss": 1.0757, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 5.724299065420561, | |
| "grad_norm": 439604.0625, | |
| "learning_rate": 5.579268292682928e-06, | |
| "loss": 1.1072, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 5.841121495327103, | |
| "grad_norm": 363162.625, | |
| "learning_rate": 5.426829268292684e-06, | |
| "loss": 1.0625, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 5.957943925233645, | |
| "grad_norm": 481210.59375, | |
| "learning_rate": 5.27439024390244e-06, | |
| "loss": 1.0626, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6890547263681592, | |
| "eval_loss": 0.9663113355636597, | |
| "eval_runtime": 14.7419, | |
| "eval_samples_per_second": 81.807, | |
| "eval_steps_per_second": 2.578, | |
| "step": 1284 | |
| }, | |
| { | |
| "epoch": 6.074766355140187, | |
| "grad_norm": 537217.875, | |
| "learning_rate": 5.121951219512195e-06, | |
| "loss": 1.0343, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 6.191588785046729, | |
| "grad_norm": 622598.75, | |
| "learning_rate": 4.9695121951219515e-06, | |
| "loss": 1.0544, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 6.308411214953271, | |
| "grad_norm": 461880.90625, | |
| "learning_rate": 4.817073170731708e-06, | |
| "loss": 1.0571, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 6.425233644859813, | |
| "grad_norm": 483674.78125, | |
| "learning_rate": 4.664634146341464e-06, | |
| "loss": 1.0273, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 6.542056074766355, | |
| "grad_norm": 507309.1875, | |
| "learning_rate": 4.51219512195122e-06, | |
| "loss": 1.0243, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 6.658878504672897, | |
| "grad_norm": 572243.0, | |
| "learning_rate": 4.359756097560976e-06, | |
| "loss": 1.0237, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 6.775700934579439, | |
| "grad_norm": 447492.46875, | |
| "learning_rate": 4.207317073170732e-06, | |
| "loss": 1.0387, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 6.892523364485982, | |
| "grad_norm": 471134.8125, | |
| "learning_rate": 4.054878048780488e-06, | |
| "loss": 1.0039, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.6923714759535655, | |
| "eval_loss": 0.9811031818389893, | |
| "eval_runtime": 14.7461, | |
| "eval_samples_per_second": 81.784, | |
| "eval_steps_per_second": 2.577, | |
| "step": 1498 | |
| }, | |
| { | |
| "epoch": 7.009345794392523, | |
| "grad_norm": 484021.0, | |
| "learning_rate": 3.902439024390244e-06, | |
| "loss": 1.0382, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 7.126168224299065, | |
| "grad_norm": 545706.5, | |
| "learning_rate": 3.7500000000000005e-06, | |
| "loss": 1.0079, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 7.242990654205608, | |
| "grad_norm": 385661.21875, | |
| "learning_rate": 3.5975609756097562e-06, | |
| "loss": 1.0593, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 7.359813084112149, | |
| "grad_norm": 389834.625, | |
| "learning_rate": 3.4451219512195124e-06, | |
| "loss": 1.0175, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 7.4766355140186915, | |
| "grad_norm": 565496.25, | |
| "learning_rate": 3.292682926829269e-06, | |
| "loss": 1.0171, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 7.593457943925234, | |
| "grad_norm": 588504.4375, | |
| "learning_rate": 3.1402439024390246e-06, | |
| "loss": 1.0401, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 7.710280373831775, | |
| "grad_norm": 439602.84375, | |
| "learning_rate": 2.9878048780487808e-06, | |
| "loss": 1.0515, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 7.827102803738318, | |
| "grad_norm": 750778.125, | |
| "learning_rate": 2.8353658536585365e-06, | |
| "loss": 1.0046, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 7.94392523364486, | |
| "grad_norm": 399670.3125, | |
| "learning_rate": 2.682926829268293e-06, | |
| "loss": 1.0062, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7205638474295191, | |
| "eval_loss": 0.9383352398872375, | |
| "eval_runtime": 15.2156, | |
| "eval_samples_per_second": 79.261, | |
| "eval_steps_per_second": 2.497, | |
| "step": 1712 | |
| }, | |
| { | |
| "epoch": 8.060747663551401, | |
| "grad_norm": 632884.5, | |
| "learning_rate": 2.530487804878049e-06, | |
| "loss": 0.9951, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 8.177570093457945, | |
| "grad_norm": 539285.375, | |
| "learning_rate": 2.378048780487805e-06, | |
| "loss": 0.9834, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 8.294392523364486, | |
| "grad_norm": 564263.875, | |
| "learning_rate": 2.225609756097561e-06, | |
| "loss": 1.0018, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 8.411214953271028, | |
| "grad_norm": 510646.3125, | |
| "learning_rate": 2.073170731707317e-06, | |
| "loss": 0.9775, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 8.52803738317757, | |
| "grad_norm": 443133.09375, | |
| "learning_rate": 1.9207317073170733e-06, | |
| "loss": 1.0046, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 8.644859813084112, | |
| "grad_norm": 459672.0625, | |
| "learning_rate": 1.7682926829268294e-06, | |
| "loss": 0.9931, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 8.761682242990654, | |
| "grad_norm": 494908.71875, | |
| "learning_rate": 1.6158536585365855e-06, | |
| "loss": 0.9988, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 8.878504672897197, | |
| "grad_norm": 472669.21875, | |
| "learning_rate": 1.4634146341463414e-06, | |
| "loss": 0.9798, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 8.995327102803738, | |
| "grad_norm": 473329.9375, | |
| "learning_rate": 1.3109756097560978e-06, | |
| "loss": 0.9948, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.7131011608623549, | |
| "eval_loss": 0.9358813166618347, | |
| "eval_runtime": 14.7963, | |
| "eval_samples_per_second": 81.507, | |
| "eval_steps_per_second": 2.568, | |
| "step": 1926 | |
| }, | |
| { | |
| "epoch": 9.11214953271028, | |
| "grad_norm": 530739.5625, | |
| "learning_rate": 1.158536585365854e-06, | |
| "loss": 0.966, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 9.228971962616823, | |
| "grad_norm": 483324.6875, | |
| "learning_rate": 1.0060975609756098e-06, | |
| "loss": 1.015, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 9.345794392523365, | |
| "grad_norm": 518854.625, | |
| "learning_rate": 8.53658536585366e-07, | |
| "loss": 0.985, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 9.462616822429906, | |
| "grad_norm": 473431.46875, | |
| "learning_rate": 7.012195121951221e-07, | |
| "loss": 0.9399, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 9.57943925233645, | |
| "grad_norm": 576602.0625, | |
| "learning_rate": 5.487804878048781e-07, | |
| "loss": 1.0286, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 9.69626168224299, | |
| "grad_norm": 585901.25, | |
| "learning_rate": 3.963414634146342e-07, | |
| "loss": 1.0004, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 9.813084112149532, | |
| "grad_norm": 543298.6875, | |
| "learning_rate": 2.439024390243903e-07, | |
| "loss": 0.9931, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 9.929906542056075, | |
| "grad_norm": 524303.25, | |
| "learning_rate": 9.146341463414634e-08, | |
| "loss": 0.9825, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7072968490878938, | |
| "eval_loss": 0.9387073516845703, | |
| "eval_runtime": 14.9995, | |
| "eval_samples_per_second": 80.403, | |
| "eval_steps_per_second": 2.533, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 2140, | |
| "total_flos": 5.510586115727032e+19, | |
| "train_loss": 1.1433691055975228, | |
| "train_runtime": 1926.7329, | |
| "train_samples_per_second": 35.449, | |
| "train_steps_per_second": 1.111 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 2140, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.510586115727032e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |